Machine Learning for Predicting Long-Term Cardiovascular Outcomes in Kidney Transplant Recipients
Authors' conclusion
ML models are feasible for predicting cardiovascular events in KTRs.
Study context
- Population
- Adult kidney transplant recipients, single centre, 2015–2024102 with MACE
- Participants
- 518 recipients
- Exposure
- CatBoost machine learning (ML) model vs multivariable logistic regression vs Soveri risk score
- Outcome
- Prediction of MACE
- Follow-up
- Mean 5.3 ± 2.3 years
Methodological detailssingle-centre, machine learning prediction model development
- Single-centre
- Machine learning prediction model development
Main finding
As reported in the abstract
MACE prediction
| Group | AUROC | Area under precision-recall curve |
|---|---|---|
| CatBoost | 0.72 (95% CI: 0.607-0.811) | 0.36 (95% CI: 0.230-0.540) |
| Logistic regression | 0.67 (95% CI: 0.559-0.772) | 0.31 (95% CI: 0.228-0.553) |
AUROC
CatBoost0.72 (95% CI: 0.607-0.811)
Logistic regression0.67 (95% CI: 0.559-0.772)
Area under precision-recall curve
CatBoost0.36 (95% CI: 0.230-0.540)
Logistic regression0.31 (95% CI: 0.228-0.553)
Observational data: associations, not effects.
Further findings
Top predictive variables
- CAD, left atrial reservoir strain, dialysis duration, LV end-diastolic volume index
Original publication
Machine Learning for Predicting Long-Term Cardiovascular Outcomes in Kidney Transplant Recipients. JACC Adv. Published 18 November 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →